2-1.Expectations relating to future laboratory automation systems:the way laboratory automation systems should be

LAS is an abbreviation of laboratory automation system.
In recent years, LAS has been the focus of considerable attention as different countries race to introduce their own LAS.
LAS is designed to effectively process everyday clinical tests and to actualize the ultimate goal of such tests, to perform them faster and more accurately, while requiring fewer specimens. Basically, LAS's goal is to incorporate automatic analyzers, automatic sample transportation and computers to manage tests systematically.
To discuss the future development of LAS, the history of LAS must be described first. Then, after discussing LAS trends around the world and presenting the problems associated with the construction of LAS, the future direction of LAS will be discussed.

In the early part of 1960 in the U.S., Dr. Skeggs found that predetermined amounts of specimens or reagents could be collected using a narrow resin pipe and a drawing pump. Consequently, he designed a revolutionary automatic analyzer that mixed a specimen and a regent in a narrow tube for chemical reactions, and this reaction solution was then sent to a cuvette connected with a colorimeter for colorimetry. Analyzers that were constructed in this method were called auto analyzers (A.A.), and they began to gain popularity around the world. This was the beginning of the automation of laboratory analyses. Naturally, these analyzers were protected by patents, and other makers could not manufacture similar products. Furthermore, even though they could only perform one type of test, each analyzer was sold for about \2 to \3 million in Japan. As a result, Japanese manufacturers were forced to sit back and watch the sale of auto analyzers skyrocket.
Then, in the 1970's, Japanese instrument manufacturers began producing automatic analyzers that could measure simultaneously perform several clinical chemical tests not requiring deproteinization by utilizing multiple reaction cells. The previous tube drawing method was renamed the continuous flow method, and the latter method was named the discrete method. After this, the central force of development and manufacture for discrete-type automatic analyzers gradually moved to Japan. The numbers of new automatic analyzers therefore increased exponentially, and Japan evolved into a leading automatic analyzer manufacturer.

When these discrete analyzers began to appear, computer technologies were also advancing rapidly, as computers were introduced into clinical laboratories in record numbers during the 1970's and 1980's. These computers were not only used to control automatic analyzers but also to organize the laboratory test requests and print reports of the laboratory data.
During this time, laboratory systematization via computers was first proposed, and the term "Clinical Laboratory System" was born. The size of automatic analyzers was increased even as, computers were getting smaller and smaller and more powerful. Furthermore, in the latter part of the 1980's, personal computers appeared and were rapidly assimilated into a variety of fields. The computer age had truly arrived.
At the same time, many clinical laboratories started to employ computer systems in which personal and mainframe computers were linked, and then between the latter part of the 1980's and the early part of the 1990's, LAN, a computer network system that linked personal computers was introduced to clinical laboratories.
To further improve the efficiency of laboratory tests, a laboratory system that directly linked a computer system to automatic analyzers was developed and became known as a Laboratory Automation System. This is the background why laboratory automation systems (LAS) has been developed.

In the early part of the 1980's, when the Kochi Medical School was opened, we had been tormented by the Law for the Total Number of Civil Servants in Japan. The Law created a persistent shortage of medical laboratory technologists and ultimately contributed to the lowering of laboratory service standards at many national medical school hospitals. As a result, a movement toward complete automatization of laboratory procedures (handling test requests, performing analyses, editing tests results from analyzers and sending information back to physicians) was introduced to compensate for the lack of personnel by incorporating an automatic specimen transportation system into an automatic analyzer and computer at Kochi Medical School. In other words, commercially available analyzers were modified, and the automatic transportation of specimens to the analyzer was realized through the use of the conveyor belts, so called the Belt Line System.
Due to the development of communication software to link automatic analyzers with the laboratory test requests online, it is now possible to collect output from analyzers and to print the test results for each patient. In this manner, complete automatization was achieved by integrating automatic analyzers, automatic specimen transportation units and computers.
Not only is this method effective in saving manpower, it makes analyzing tests more reliable and faster, while requiring fewer specimens. Furthermore, since workers have fewer opportunities to come into direct contact with specimens, they are less likely to be exposed to harmful pathogens.

As stated above, due to the advances in automatic analyzers, computers and transportation units, the complete automatization of clinical laboratories has expanded to include all services in a clinical laboratory of Japan.
Due to chronic shortages in manpower, national medical schools requested funds from the Ministry of Education in order to undertake LAS construction. The year was 1990, and the Japanese economy was at its peak, so these budgets were readily approved. Consequently, out of 43 national medical school hospitals, LAS was installed at about 90% of the clinical laboratories.
This was the beginning of the problem.
As each institution was competing to install LAS, makers tried to construct a system utilizing their own analyzers. However, no one Japanese maker could provide all analyzers used in a clinical laboratory. As a result, each clinical laboratory began to search for machines that were compatible with a transportation system, but as the structure of analyzers and sample transportation racks varied greatly, it was extremely difficult to find system combinations that were required by users. Therefore, at some clinical laboratories, complete laboratory automatization was forcibly achieved using robots to connect automatic specimens transportation system without considering expense.
Although makers were able to install automated systems, since they did not thoroughly understand the workings of a clinical laboratory, some laboratory automation systems had serious shortcomings. The online system of one laboratory could only handle communications within the laboratory, but the laboratory test requests or test result printouts were not integrated. Some laboratories automated laboratory services, but were still forced to hire more workers. As other clinical laboratories employed systems without questioning their manufacturer, it was necessary for them to increase the number of specimen containers, as well as the volume of blood samples collected from patients.
Yet in the worst cases, because a clinical laboratory was completely managed by a computer, it was impossible to intervene manually in the laboratory tests, and since all tests must be registered with the computer, emergency tests could not be performed until already-in-progress routine tests were completed. Hence, urgent tests could not be performed immediately at some clinical laboratories. One laboratory had to purchase analyzers separately for emergency purposes in order to correct the shortcomings of existing laboratory automation systems. Moreover, when a physician questions the outcome of a test, some systems can not identify the type and location of the cause of test anomalies.
Despite these problems, most clinical laboratories report that the speed of clinical laboratory analyses has clearly improved, and many users have voiced the opinion that LAS has also contributed to the improvement of test accuracy. Furthermore, depending on the design of a laboratory automation system, the number of technologists performing routine tests has decreased in many clinical laboratories in Japan.

Since the latter half of 1990, calls for complete automation systems have come from all over the world.
There was already a movement toward complete automation systems in other countries before 1990. Mayo clinic in Rochester, Minnesota, attempted to install a complete automation system in 1988, and there were several meetings between the clinic and a Japanese maker, which ultimately fell through. One reason was high costs, and the other was that the maker had determined that it would be difficult to provide sufficient after-care service because of the physical distance between the two countries.
Several other clinical laboratories in the United States actively attempted to install a Japanese automation system, but due to the above reasons, none of the deals were finalized until, in January of this year, Dr. Teplitz at Beth Israel Hospital in New York successfully installed a transportation system manufactured by IDS in Japan through Coulter Corp. in the U.S. Shortly after, an entire laboratory system designed by Hitachi, Ltd. was installed in the Medical Foundation Laboratory Center in South Bend, Indiana.
In March of this year, I was pleasantly surprised when I was invited to visit the Sam Sung Medical Center in Seoul, South Korea. Laboratory systems produced by Japanese firms, IDS and Sysmex, have been operated since last November by 85 technologists, and was handling not only chromosome analyses and gene analyses such as DNA tests, but also general tests and trace metal analyses. This was an ideal example of how the laboratory automation system can function within a medical school hospital clinical laboratory.
LAS has been continuously implemented in the U.S. and South Korea in the past few years, trend that will likely continue to spread across the globe in the next few years.

As stated above, LAS is continuously expanding throughout the world, and many predict LAS will be installed in most clinical laboratories by the year 2010.
At present, every laboratory that considers implementing LAS is astonished by the cost. To get more low costs, the "ideal" LAS plan has to be modified and unnecessary component from the original plan of the automation system removed in most cases.
The next step in cost-cutting is the downsizing of computer-related components instead of analyzers, since lowering the variety and number of analyzers from the ideal combination reduces the number of automatically analyzed items and negatively affects test efficiency. As a result, in this economic approach to systematization, overall system slow down is unavoidable since the computer's processing speed slows during peak hours. Therefore, saving money on computer related components results in compromises in processing capacity.
Therefore, an LAS should only be planned and designed by actual laboratory workers. It is often the case that a maker plans and designs LAS, and the laboratory workers are only concerned about paying for the system. This is unwise, as manufacturers by nature will attempt to cut costs and increase profits by reducing expenses such as by utilizing existing software that was designed for older systems.
If you plan and design a system yourself, however, then you can customize the system to your needs. In addition, the design process will also be a way to gain knowledge by trial and error, and this knowledge will be valuable for revising the system in the future. Furthermore, I believe that by constantly thinking about your system, and how it applies to your unique needs, new ideas will come to light, ideas that will further the development of LAS.

When designing a laboratory system, one should first list all necessary analyzers and then sketch a chart as you would arrange building blocks, so that an entire system can be viewed.
Even when a desired system is over budget, the overall flow of an ideal system should be drawn by considering matters such as the flow of specimens, communication, specimen reception, specimen storage and report output. Next, based on the sketch, a final drawing should be made, and absolutely necessary components identified on the drawing. Components that are required for the initial phase are then listed, and their total cost is calculated and compared with the proposed budget. If the list is under the budget, then you can add more components. If the list is over the budget, then you will need to reconsider which components are absolutely necessary for the initial phase.
Even if an entire system can be paid for at once, it is better to construct the system in at least two stages, and the budget should be distributed over a period of several years to minimize risk. In other words, the overall system diagram should be drawn first, and the system should be constructed utilizing only half of the budget at first. It is recommended that the remaining analyzers be purchased one by one over a period of several years to complete the system. The most important thing is to actually use the system to get a grasp of its defects and shortcomings. These problem points can be corrected with funds from subsequent years.
Nevertheless, due to the budgeting process used in national medical school in Japan, it will be very difficult to follow the above suggestion. Hence, it will be necessary to negotiate with the accounts section, especially the business affairs departments, so that funds can be secured over a period of several years. In many cases, even when a budget is set, orders are placed in September or October, and due to budget constraints, all components must be delivered by the end of March in Japan.
Quite frankly, it is absolutely impossible to construct a perfect system in six months. However, public clinical laboratories such as national medical school hospitals clinical laboratories must set up systems even if they are initially insufficient or incomplete. As a result of the above factors, in many clinical laboratories, computer software is never quite complete and up to date. Furthermore, if a problem arises more than one year after delivery, the full warranty does not apply, and if there is not enough money available to correct the software, the entire system is compromised. To resolve these problems, I strongly recommend that systematization proceed in at least two stages.

A good, effective system can not be constructed in a day. Like good paintings, effective systems are designed based on years of experience and with unique talents and with unique insights. Even a very talented artist can not create new ideas if he or she is not intimate with the subject. So good laboratory automation systems are conceptualized through years of experience and accumulated ideas. Through trial and error, new ideas are realized, and systems are continuously improved.
The same applies to automatic analyzers. Numerous trial models have been produced, and by learning from the many mistakes that were made on the way, today's accurate analyzers were developed to the point that they require only micro amounts of specimens. Even though specimen transportation systems were first developed about 16 years ago based on transportation systems that are utilized in other fields, they have advanced rapidly in a short period of time due to the extensive efforts of all involved.
Furthermore, the function and performance of computers used in clinical laboratories have improved more than ten fold in recent years.
Constructing an ideal laboratory automation system by skillfully incorporating these components is indeed akin to creating a piece of art.

The ultimate goal of the laboratory tests is to derive the analyzed results quickly and accurately, and then reliably deliver the results to a client of physician in order to aid clinical treatment of diseases.
As a means to achieve this goal, with years of clinical laboratory experience, a concept of LAS was developed by incorporating analyzers, specimen transportation systems and computers that are being developed in the rapidly advancing fields of mechanical and electronic engineering. It is possible that superior methods will be developed in the future. Nonetheless, it is unequivocally clear that the present method is one way to achieve an ideal laboratory.
Therefore, LAS should not be a profit source for the analyzer industry but remain first and foremost a technique used to strive for the realization of an ideal laboratory, and an example of how human ingenuity can contribute to the advancement of medicine.

Masahide Sasaki, M.D.
Professor of Laboratory Medicine, Kochi Medical School
Place of Birth/ Yamaguchi, Japan
Professional Experience/
1961: Graduated from Yamaguchi Medical School
1965: Worked as an internist for the Hiroshima Atomic Bomb Causality
Committee (ABCC)
1967: Served as a chief of Clinical chemistry at Kawasaki Hospital
1970: Studied abroad at Michael Reese Hospital in Chicago for two years
1972: Returned to Japan as a assistant professor of internal medicine at
Kawasaki Medical School
1976: Served as a professor of Laboratory Diagnosis at Kawasaki Medical
School and as a vice president of Kawasaki Paramedical College
1981: Served as a professor and a Director of Department of the Clinical
Laboratory at Kochi Medical School
1989: Served as a professor of the Department of Clinical Laboratory
Medicine at Kochi Medical School